Antioxidant effects of photodegradation product of nifedipine

Article abstract of DOI:10.1248/cpb.59.208

Recently, increasing evidence suggests that the antihypertensive drug nifedipine acts as a protective agent for endothelial cells, and that the activity is unrelated to its calcium channel blocking. Nifedipine is unstable under light and reportedly decomp

Full text of DOI:10.1248/cpb.59.208

2
08
Regular Article
Chem. Pharm. Bull. 59(2) 208—214 (2011)
Vol. 59, No. 2
Antioxidant Effects of Photodegradation Product of Nifedipine
a
b
b
a
a
Yuya HORINOUCHI, Koichiro TSUCHIYA, Chiaki TAOKA, Soichiro TAJIMA, Yoshitaka KIHIRA,
b
c
c
d
e
Yuko MATSUDA, Kozo SHISHIDO, Masahiro YOSHIDA, Shuichi HAMANO, Kazuyoshi KAWAZOE,
Yasumasa IKEDA, Keisuke ISHIZAWA, Shuhei TOMITA, and Toshiaki TAMAKI*
a
a
a
,a
a
Department of Pharmacology, The Institute of Health Bioscience, The University of Tokushima Graduate School;
Department of Medical Pharmacology, The Institute of Health Bioscience, The University of Tokushima Graduate School;
b
c
Department of Synthetic Organic Chemistry, The Institute of Health Bioscience, The University of Tokushima Graduate
d
School; Department of Pathological Science and Technology, The Institute of Health Bioscience, The University of
Tokushima Graduate School; 3–18–15 Kuramoto, Tokushima 770–8503, Japan: and Department of Pharmacy, Tokushima
e
University Hospital; 3–18–15 Kuramoto, Tokushima 770–8503, Japan.
Received August 30, 2010; accepted November 16, 2010; published online November 17, 2010
Recently, increasing evidence suggests that the antihypertensive drug nifedipine acts as a protective agent
for endothelial cells, and that the activity is unrelated to its calcium channel blocking. Nifedipine is unstable
under light and reportedly decomposes to a stable nitrosonifedipine (NO-NIF). NO-NIF has no antihypertensive
effect, and it has been recognized as a contaminant of nifedipine. The present study for the first time demon-
strated that NO-NIF changed to a NO-NIF radical in a time-dependent manner when it interacted with human
umbilical vein endothelial cells (HUVECs). The electron paramagnetic resonance (EPR) signal of NO-NIF radi-
cals in HUVECs showed an asymmetric pattern suggesting that the radicals were located in the membrane. The
NO-NIF radicals had radical scavenging activity for 1,1-diphenyl-2-picrylhydrazyl, whereas neither NO-NIF nor
nifedipine did. In addition, the NO-NIF radical more effectively quenched lipid peroxides than NO-NIF or
nifedipine. Furthermore, NO-NIF attenuated the superoxide-derived free radicals in HUVECs stimulated with
LY83583, and suppressed iron-nitrilotriacetic acid (Fe-NTA)-induced cytotoxicity in rat pheochromocytoma
(PC12) cells. Our findings suggest that NO-NIF is a candidate for a new class of antioxidative drugs that protect
cells against oxidative stress.
Key words nifedipine; nitrosonifedipine; antioxidant; electron paramagnetic resonance; reactive oxygen species; endothelial
cell
Calcium channel blockers have been used for more than al. reported that NO-NIF scavenged 2,2ꢀ-azo-bis(2-amidino-
30 years to treat a variety of cardiovascular diseases includ- propane)-derived alkylperoxyl (ABAP) radicals and that the
ing angina, arrhythmias, and hypertension. Nifedipine [1,4- activity was 2.3 times greater than that of trolox, a water-sol-
27)
dihydro-2,6-dimethyl-4-(2-nitrophenyl)-3,5-pyridinedicar- uble vitamin E analogue. In addition, Mi sˇ ík et al. postu-
boxylic acid dimethyl ester] is a 1,4-dihydropyridine-deriva- lated that NO-NIF interacts with unsaturated lipids to form
tive calcium channel blocker widely used for treatment of hy- NO-NIF radicals that are responsible for the antioxidant
28)
pertension and angina. Over the past decade, many clinical properties of NO-NIF. We therefore hypothesized that the
studies have suggested that nifedipine has not only hypoten- antioxidant activity of NO-NIF is a critical element in the
sive effects but also cardiovascular organ-protective effects in organ-protective effects of nifedipine. Therefore, in this
1—4)
patients with heart and circulatory diseases.
In addition, study, we investigated the ability of NO-NIF to protect cells
it has been reported that nifedipine ameliorates endothelial from oxidative stress using cultured human umbilical vein
5—9)
10,11)
dysfunctions
through an anti-apoptotic effect
in endothelial cells (HUVECs) and rat pheochromocytoma
human endothelial cells. However, it is unlikely that nifedip- (PC12) cells.
ine exerts these effects through its calcium-blocking property
because endothelial cells do not have voltage-dependent L- Experimental
type calcium channels, which are thought to be the target of
Materials Nifedipine, methanol, 1,1-diphenyl-2-picrylhydrazyl (DPPH),
palmitic acid, linoleic acid, iron(III) nitrate nonahydrate(Fe(NO ) ·
12)
1,4-dihydropyridine derivatives. The antioxidant activity
3 3
9
H O), 4-hydroxy-2,2,6,6-tetramethylpiperidinyloxy (Tempol), superoxide
2
may be one possible mechanism responsible for the organ-
protective effects of 1,4-dihydropyridine calcium channel
dismutase (SOD; from bovine erythrocytes), and nitrilotriacetic sodium
NTA-Na) were purchased from Wako Pure Chemical Industries (Tokyo,
(
13—15)
blockers.
However, the antioxidant activity of nifedipine Japan). Linolenic acid was obtained from Tokyo Chemical Industries
(Tokyo, Japan). Arachidonic acid, eicosapentaenoic acid (EPA), and docosa-
is reported to be less prominent than that of other 1,4-dihy-
13,14)
hexaenoic acid (DHA) were from Cayman Chemicals (Ann Arbor, MI,
U.S.A.). Stearic acid was from MP Biomedicals, Inc. (OH, U.S.A.). 5,5-Di-
methyl-1-pyrroline-N-oxide (DMPO) was purchased from Labotec (Tokyo,
Japan). LY83583 was obtained from Calbiochem (Darmstadt, Germany). All
dropyridine calcium channel blockers.
Nifedipine is extremely sensitive to light and can be con-
verted to its nitroso analog, nitrosonifedipine [2,6-dimethyl-
4
-(2-nitrosophenyl)-3,5-pyridinedicarboxylic acid dimethyl other reagents were of analytical grade. Water was demineralized and further
16—22)
purified using a Milli-Q system from Millipore (Bedford, MA, U.S.A.). The
stock solution of 5 mM iron-nitrilotriacetic acid (Fe-NTA) was prepared by
ester] (NO-NIF), under normal room illumination.
On
the other hand, NO-NIF is also enzymatically produced from
mixing 10 ml of 10 mM Fe(NO ) ·9H O with 10 ml of 20 mM NTA-Na im-
mediately before use.
23)
3
3
2
nifedipine without exposure to light. The ability of NO-
NIF to block calcium channels is quite weak or non-existent
29)
Preparation of NO-NIF NO-NIF was prepared according to a previous
24—26)
19)
compared with that of nifedipine.
Meanwhile, Yanez et report on nifedipine. Briefly, 500 ml of nifedipine solution (10 mM) in
∗
To whom correspondence should be addressed. e-mail: tamaki@basic.med.tokushima-u.ac.jp
© 2011 Pharmaceutical Society of Japan

February 2011
209
methanol was placed in glass beaker and then exposed to a halogen light
Plus, Bruker Bio Spin GmbH, Rheinstetten, Germany) with an X-band cav-
ity (ER4119HS, Bruker) to collect all EPR spectra. All EPR signals were
obtained at 9.5 GHz with 100 kHz modulation. Samples were transferred to
glass capillary tubes (10 ml, Drummond Co., Broomall, PA, U.S.A.) and set
into the EPR cavity for the measurements. The spin concentration of NO-
NIF-derived radicals was determined by the double integration using Bruker
WinEPR software, and radical concentrations were calculated by reference
to the double integral of signals from a known concentration of the freshly-
prepared stable radical Tempol (0—0.1 mM) run under identical conditions,
(
500 W, Kodak Ektagraphic III Projector, Kodak, Rochester, NY, U.S.A.)
ꢁ
2
with constant stirring. The UVA intensity of the source was 1.0 mW cm
when measured with a UV radiometer UVR-3036/S (Topcom, Tokyo, Japan)
at the position of the sample. Every 2 h a sample was removed and subjected
to HPLC with UV detection. HPLC was performed using a JASCO 880-PU
pump (Tokyo, Japan) and a manual injector with a 20 ml loop. The separa-
tion was carried out on a Cosmosil C -AR-II (4.6ꢂ100 mm) column. The
5
18
mobile phase consisted of a mixture of methanol (60%, v/v) and water fil-
tered through a 0.45 mm filter (Millipore). HPLC analysis was performed at
room temperature under isocratic conditions with a flow rate of 0.7 ml/min,
then monitored using a UV spectrum detector (Hitachi, L-7400, Tokyo,
30)
as described previously. HUVEC-derived ROS formation was studied by
DMPO spin trapping using the methods of Souchard et al. with slight modi-
31)
fications. Instrument conditions are described in each figure legend.
Japan) at 292 nm. The eluent corresponding to newly observed peaks was
Leucomethylene Blue Assay The leucomethylene blue (LMB)
1
32,33)
collected, evaporated for determination of its structure by H-NMR assay
is based on the hemoglobin-catalyzed oxidation of colorless ben-
zoyl leucomethylene blue to detect the existence of lipid hydroperoxide. Re-
duction of the lipid hydroperoxide by an antioxidant to the corresponding al-
cohol results in a reduction of LMB oxidation, which is an indication of
anti-peroxide activity. The LMB assay was performed in 96-well microtiter
plates. Sample solutions were mixed with LMB solution (5 mg N-benzoyl
leucomethylene blue in 8 ml dimethylformamide), then 100 ml of the solu-
tion was mixed with 50 ml of S-13-hydroperoxyoctadecadienoic acid (13-
HPODE; 240 mM in PBS containing 5% EtOH) and 100 ml of the catalytic
reagent (50 mM potassium phosphate buffer pH 5.0 containing 1.4% Triton
X-100 and 5.5 mg hemoglobin) in the microtiter plate. After 10 min incuba-
13
(
400 MHz) and C-NMR (100 MHz) in CDCl solution, and referenced to
3
tetramethylsilane (TMS) (0.00 ppm) using a JEOL GSX400 spectrometer
Tokyo, Japan). IR spectra were determined by a JASCO FT/IR-420 spec-
trometer (Tokyo, Japan), melting point by a Yanaco MP-500D apparatus
Kyoto, Japan), and a mass spectrometry by a Waters Micro Mass LCT-Pre-
mier spectrometer (Japan Waters, Tokyo, Japan). The data were as follows;
(
(
ꢁ
1
1
mp 94.2—94.6 °C; IR (KBr) 1730, 1558, 1493 cm ; H-NMR (400 MHz,
CDCl ) d: 2.67 (s, 6H), 3.39 (s, 6H), 6.55 (dd, Jꢃ8.0, 1.2 Hz, 1H), 7.44 (dt,
3
Jꢃ8.0, 1.2 Hz, 1H), 7.52 (dd, Jꢃ8.0, 1.2 Hz, 1H), 7.72 (dd, Jꢃ8.0, 1.2 Hz,
1
3
1
H); C-NMR (100 MHz, CDCl ) d: 23.4ꢂ2 (CH ), 51.9ꢂ2 (CH ), 107.7
3
3
3
(
(
(
[
CH), 127.7 (Cq), 128.8 (CH), 130.6 (CH), 135.0 (CH), 139.9 (Cq), 144.3 tion at room temperature, absorbance at 660 nm was monitored using a mi-
Cq), 156.4 (Cq), 161.6 (Cq), 167.6 (Cq); high resolution-mass spectrometry
HR-MS) (electrospray ionization (ESI)) m/z Calcd for C H N O
M ꢄH ] 329.1137.
After 18 h irradiation under our experimental conditions, nifedipine had
crotiter plate reader.
DPPH Radical Assay The antioxidant activities of NO-NIF in the pres-
ence or absence of fatty acids were assayed using the 1,1-diphenyl-2-picryl-
1
7
17
2
5
ꢄ
ꢄ
34)
hydrazyl (DPPH) radical dismutation technique with modification. Briefly,
100 ml of DPPH methanol solution (0.5 mM) was diluted with 300 ml of
methanol, then mixed with 80 ml of buffer solution (50 mM acetate buffer,
pH 5.5) and 20 ml of sample solution diluted by methanol. After 30 min in-
been completely converted to NO-NIF with a purity of more than 99%.
Then, the reaction mixture was evaporated nearly to dryness and recrystal-
lized several times from methanol.
Cell Culture HUVECs (Takara Bio, Shiga, Japan) were cultured at a cubation, the reaction mixture was transferred to the EPR quartz flat cell,
5
density of 1ꢂ10 cells/ml in 60 mm dishes (Iwaki Glass Co., Ltd.) in En-
and the EPR spectrum of DPPH radicals was measured. The relative concen-
tration of DPPH radicals was obtained by double integration of each spec-
dothelial Basal Medium 2 (EBM-2, Takara Bio) supplemented with 10%
fetal bovine serum (FBS), gentamicin sulfate (50 mg/ml), and amphotericin- trum. EPR spectrometer settings are given in the figure captions.
B (50 mg/ml) in addition to human fibroblast growth factor B (10 ng/ml), re-
combinant human epidermal growth factor (20 ng/ml), recombinant human
vascular endothelial growth factor (rhVEGF; 1 ng/ml), insulin-like growth
factor 1 (IGF-1; 1 ng/ml), ascorbic acid (1 mg/ml), heparin (3 ng/ml), and hy-
Statistical Analysis Values are expressed as meansꢅS.D. for 3—6 sep-
arate experiments. One-way analysis of variance was used to determine sig-
nificance among groups, after which a modified t-test with the Bonferroni
correction were used for comparison between groups. Values of pꢆ0.05
were accepted as statistically significant.
drocortisone (0.4 mg/ml) at 37 °C in an incubator containing 5% CO . After
2
2
d in culture, HUVECs were washed with prewarmed (37 °C) phosphate-
buffered saline (PBS) without Ca/Mg. The cells were harvested by
trypsinization and washed first in cold complete medium and subsequently
in PBS with Ca/Mg (PBS(ꢄ)), then suspended at cell density of
5
1
ꢂ10 cells/ml in PBS(ꢄ).
Rat pheochromocytoma PC12 cells were inoculated at a density of
5
1
ꢂ10 cells/ml in 24-well plastic plates (Iwaki). Each well contained 600 ml
of RPMI 1640 (Sigma, St. Louis, MO, U.S.A.) supplemented with 10%
FBS, and cells were cultivated at 37 °C in a humidified atmosphere contain-
ing 5% CO for 24 h, then starved in RPMI 1640 without FBS for 16 h prior
2
to experiments.
Lactate Dehydrogenase Assay The cell damage induced by oxidative
stress and its attenuation by NO-NIF was evaluated by a lactate dehydroge-
nase (LDH) assay (Cytotoxicity Detection kit, Roche, Grenzach-Wyhlen,
Germany). Briefly, the RPMI 1640 medium was aspirated from PC12 cells,
and the plates were washed twice with 2—3 ml of Krebs–Ringer Hepes
(
KRH) buffer (NaCl 125 mM, KCl 4.7 mM, CaCl2 2.2 mM, N-(2-hydroxy-
ethyl)piperazine-Nꢀ-2-ethanesulfonic acid (HEPES) 10 mM, KH PO 1.2 mM,
2
4
MgSO 1.2 mM, and glucose 6 mM, pH 7.4). The PC12 cells were then incu-
4
bated in KRH buffer containing either 10 mM NO-NIF or 10 mM nifedipine
for 30 min at 37 °C in a dark environment. The KRH buffer was then re-
moved, and the cells were oxidized by the addition of Fe-NTA complex dis-
solved in KRH buffer (10 mM) and then incubated for 4 h at 37 °C. The plates
were centrifuged at 400 g and 4 °C for 5 min, and 50 ml aliquots were taken
to quantify the LDH. The assay was performed according to the manufac-
turer’s instructions. The release of intracellular LDH to the extracellular
medium was measured by determining the enzyme activity and expressed as
a percentage of total cellular activity. The absorbance was measured at
Fig. 1. Effect of Light on Nifepedine
490 nm using a plate reader. We chose Fe-NTA complex as the oxidant be-
ꢁ
2
Nifedpine (10 mM) methanol solution was exposed to a halogen light (1.0 mW cm ),
and aliquots were sampled at 0 (A), 7 h (B), and 18 h (C). Open and closed circles rep-
resent nifedipine and NO-NIF, respectively. The injection volume was 20 ml, and chro-
matographic peaks were monitored at 292 nm. Eluent at tꢃ3.6 was collected and then
evaporated for NMR, IR, and MS.
cause Fe-NTA is a good reagent for generation of lipid peroxidation and re-
active oxygen species (ROS).
Free Radical Analysis by EPR Spectroscopy The free radical metabo-
lites of NO-NIF were examined using a Bruker EPR spectrometer (EMX
29)

2
10
Vol. 59, No. 2
Results
generated apparent three-line EPR signals. Neither NO-NIF
Synthesis of NO-NIF When nifedipine was exposed to (Fig. 3A) nor NO-NIF with saturated fatty acids (Figs.
light, it (retention timeꢃ4.3 min) was gradually changed to 3G, H) gave EPR signals under the same conditions.
another compound (retention timeꢃ3.4 min) over time (Fig. 1).
Subsequently, we investigated the time course of NO-NIF
In order to determine the structure of the compound pro- radical generation in the presence of various concentrations
1
13
duced, we conducted H-NMR, C-NMR, IR, and MS meas- of linoleic acid. As shown in Fig. 4, the NO-NIF-derived
urements. According to the data, (Found 329.1148), it was EPR signal stimulated by linoleic acid was augmented in a
concluded that the new compound was NO-NIF as re- time-and concentration-dependent manner.
19)
ported (Fig. 1).
Reduction of DPPH Radical by NO-NIF DPPH is con-
Lipid Peroxidation In order to measure the scavenging sidered to be a model compound of a lipophilic radical, and
activity of nifedipine and NO-NIF toward lipid peroxides, we it has been used to verify the radical scavenging activity of
conducted LMB assays. This method uses the hemoglobin- various sorts of antioxidants. The DPPH radical is scavenged
catalyzed oxidation of colorless N-benzoyl leucomethylene by antioxidants through donation of a hydrogen to form the
blue to detect the presence of 13-HPODE. The mixture of stable non-radical DPPHH molecule on the basis of follow-
35)
NO-NIF with linoleic acid was prepared by mixing 5 mM ing equation : DPPH˙ꢄHX→DPPHHꢄ˙X (where HX rep-
NO-NIF with 5 mM linoleic acid and then incubated for one resents an antioxidant). The more the DPPH radical concen-
hour at 37 °C. As shown in Fig. 2, NO-NIF with linoleic acid tration decreased, the more potent the antioxidant activity of
showed more potent activity than NO-NIF alone, and NO- the compound through its hydrogen-donating property. Nei-
NIF had higher potency, depending on anti-lipid peroxida- ther NO-NIF nor NO-NIF with saturated fatty acids (Fig. 5F:
tion, than nifedipine did (pꢆ0.01).
palmitic acid and Fig. 5G: stearic acid) affected the DPPH
Free Radical Production by NO-NIF It was reported radical intensity. However, in the presence of NO-NIF with
23)
that NO-NIF interacts with oleic acid or dioleoyl phos- unsaturated fatty acids (Fig. 5C: linoleic acid, Fig. 5D:
28)
phatidylcholine to form NO-NIF radical adducts. There- linolenic acid, and Fig. 5E: arachidonic acid), the DPPH sig-
fore, we investigated whether other fatty acids that comprise nal intensity significantly decreased. Nifedipine did not af-
the cell membrane and related fatty acids generate NO-NIF fect the DPPH radical signal intensity (data not shown).
radicals or not. As shown in Figs. 3B—F, co-incubation of
NO-NIF with unsaturated fatty acids (linoleic, linolenic,
arachidonic, eicosapentaenoic, or docosahexaenoic acid)
Formation of NO-NIF Radical in HUVECs Next, we
Fig. 4. Time Course of Linoleic Acid-Induced NO-NIF Radical Formation
NO-NIF (1 mM) was mixed with linoleic acid (closed square, 0.1 mM; closed circle
1
3
mM) in methanol and kept in airtight brown glass vials for the indicated periods at
7 °C. Each EPR signal was double-integrated to obtain arbitrary units of spin concen-
Fig. 2. Effect of Various Agents on Reduction of 13-HPODE
13-HPODE (48 mM) was mixed with 250 mM of nifedipine (Nif), NO-NIF, or a mix-
tration, then compared with that of a known concentration of Tempol as described in
Experimental. Values are meansꢅS.D. of four independent experiments. EPR spec-
trometer instrument settings were power 20 mW, center field 3362 gauss, sweep width
ture of NO-NIF plus linoleic acid (1 : 1), then incubated for 10 min at room tempera-
ture. The remaining 13-HPODE was measured at 660 nm by LMB assay. Values are
meansꢅS.D. of four independent experiments. ∗∗ pꢆ0.001 vs. control, and ## pꢆ0.01
between samples.
50 gauss, sweep time 4 min, modulation frequency 100 kHz, modulation width 1 gauss,
and time constant 0.1 s.
Fig. 5. Effect of NO-NIF with or without Fatty Acids on DPPH Radical
Intensity
Fig. 3. EPR Spectra of NO-NIF with Fatty Acids
NO-NIF (A; final conc. 5 mM) and 5 mM (final conc.) fatty acid (B, linoleic acid; C,
linolenic acid; D, arachidonic acid; E, eicosapentanoic acid; F, docosahexaenoic acid;
G, palmitic acid; H, stearic acid) were mixed with methanol and incubated for 4 h at
NO-NIF (0.1 mM) was mixed with 0.1 mM fatty acids for 144 h at 37 °C, then with
0.1 mM DPPH. A, 0.1 mM DPPH; B, 0.1 mM DPPH with 0.1 mM NO-NIF; C, B with
0.1 mM linoleic acid; D, B with 0.1 mM linolenic acid; E, B with 0.1 mM arachidonic
acid; F, B with 0.1 mM palmitic acid; and G, B with 0.1 mM stearic acid. ∗ pꢆ0.05 and
∗∗ pꢆ0.01 vs. DPPH with NO-NIF(B). Data are expressed as the meansꢅS.D. of three
separate experiments. EPR spectrometer instrument settings were the same as in Fig. 3.
37 °C. The EPR spectrometer instrument settings were power 20 mW, center field 3521
gauss, sweep width 100 gauss, sweep time 4 min, modulation frequency 100 kHz, mod-
ulation width 1 gauss, and time constant 0.163 s.

February 2011
211
Fig. 8. Protective Effect of NO-NIF on Cultured PC12 Cells against Fe-
NTA-Induced Cytotoxicity
PC12 cells were pretreated with 10 mM nifedipine or 10 mM NO-NIF for 30 min be-
fore being exposed to 10 mM Fe-NTA for 4 h or were treated with Fe-NTA without any
pretreatment for 4 h as a control. Then, the release of LDH into the medium was meas-
ured as described in “Experimental.” Data are expressed as the meansꢅS.D. of three
experiments. ∗∗ pꢆ0.001 vs. control.
Fig. 6. EPR Spectra of NO-NIF-Derived Radicals in the Presence of HU-
VECs
5
HUVECs (1ꢂ10 cells/ml) were incubated in PBS(ꢄ) with 1 mM NO-NIF at 37 °C in
a dark environment. Aliquots (30 ml) of cell suspension were subjected to EPR meas-
urement at the indicated time points. EPR spectrometer instrument settings were same
as in Fig. 3 except that four scans were conducted to improve the signal/noise ratio.
5
min prior to the addition of DMPO and LY83583. When
HUVECs were incubated with 100 mM DMPO alone, no
EPR signal was observed (Fig. 7A) with either nifedipine
(
10 mM) or NO-NIF (10 mM) (Figs. 7B, C). However, after 5
min incubation of HUVECs with 100 mM DMPO, followed
by addition of 100 mM LY83583, the EPR signal typical of
N
H
DMPO/˙OH (a ꢃa ꢃ14.7 gauss) was observed (Fig. 7D) as
36)
reported previously. When SOD (100 U/ml) was co-incu-
bated with HUVECs and stimulated by LY83583, the EPR
signals of DMPO/˙OH were suppressed (Fig. 7E). NO-NIF
decreased the DMPO/˙OH adducts induced by LY83583 in a
concentration-dependent manner (Figs. 7G—I), whereas
nifedipine did not (Fig. 7F).
Effects of NO-NIF on Fe-NTA Cytotoxicity When
PC12 cells were treated with 10 mM Fe-NTA, 66.2ꢅ2.1% of
LDH was released from cells. NO-NIF treatment signifi-
cantly attenuated the cytotoxicity induced by Fe-NTA
Fig. 7. Effects of Nifedipine and NO-NIF on DMPO Spin Adducts Ob-
tained from LY83583-Stimulated HUVECs
5
HUVECs (1ꢂ10 cells/ml) were incubated in the presence of the spin-trap agent
DMPO (100 mM) and stimulated with LY83583 for ROS generation. Spectrum A: HU-
(2.5ꢅ3.6%), whereas nifedipine did not (66.2ꢅ10.8%). The
VECs with DMPO; spectrum B: same as A butꢄ10 mM nifedipine; spectrum C: same intrinsic cytotoxicity of NO-NIF (5.1ꢅ3.0%) was negligible
as A butꢄ10 mM NO-NIF; spectrum D: same as A butꢄ100 mM LY83583; spectrum E:
same as D butꢄ100 U/ml SOD; spectrum F: same as D butꢄ10 mM nifedipine; spec-
compared with that of nifedipine (19.5ꢅ8.7%) under this ex-
trum G: same as D, butꢄ1 mM NO-NIF; spectrum H: same as D butꢄ5 mM NO-NIF; perimental condition (Fig. 8).
spectrum I: same as D butꢄ10 mM NO-NIF. Closed circles represent DMPO/˙OH sig-
nals. EPR spectrometer instrument settings were the same as in Fig. 3.
Discussion
Nifedipine is one of the most widely used dihydropyridine
investigated whether the NO-NIF radical was produced dur- (DHP)-based calcium antagonists, and it has been used for
ing incubation with HUVECs. One milliliter of the cell sus- the treatment of hypertension and angina pectoris since the
pension was mixed with 1 mM NO-NIF, then incubated for up 1960s. In addition to its hypotensive and anti-anginal effects,
to 24 h at 37 °C in a dark environment for NO-NIF radical pleiotropic effects of nifedipine were recently reported: anti-
measurement. Aliquots of the suspension were transferred to apoptotic and anti-inflammatory effects on endothelial cells
6
0 ml glass capillary tubes at the times indicated in Fig. 6, that control the development and progression of atherosclero-
1,6—11,37—41)
and each EPR spectrum was measured. When HUVECs were sis.
However, it is not clear how nifedipine exerts
incubated with NO-NIF, non-symmetrical EPR signals of these pleiotropic effects because its antioxidative activity,
NO-NIF radicals were observed (Fig. 6), and the shapes of which is responsible for the cardiovascular protective ef-
1
3,14,42)
43—46)
the EPR signal were identical to those from NO-NIF with fect,
is weak in comparison to other DHPs.
liver and heart. The EPR signal reached the maximum at dentally, nifedipine is light-sensitive, and is converted com-
h, although EPR signals were still observed till the end of pletely to its nitroso analog, 2,6-di-methyl-4-(2-nitro-
observation period (24 h). sophenyl)-3,5-pyridine-carboxylic acid dimethyl ester (NO-
Effect of Nifedipine or NO-NIF on Production of Reac- NIF, Fig. 1) without further photochemical degradation under
Inci-
23)
28)
6
23)
tive Oxygen Species We previously reported that LY83583 ordinary light in 24 h. In addition, it has been reported that
generated superoxide anion radicals (O˙ ) from cultured NO-NIF was enzymatically generated from nifedipine. To
cells using the EPR-spin trapping technique. Here, we used date, it has been reported that NO-NIF reduced the alkylper-
the technique to investigate whether NO-NIF could scavenge oxyl radicals on its nitroso aromatic group, the kinetic rate
reactive oxygen species (ROS) in cultured HUVECs. constant was ten and two times higher than that of nifedipine
ꢁ
23)
2
36)
27)
Nifedipine or NO-NIF was added to the cell suspension and Trolox, respectively, and NO-NIF reacted with unsatu-

2
12
Vol. 59, No. 2
rated fatty acids to form nitroxide radicals via a pseudo- were maintained for 24 h. Because the shape of the EPR sig-
18,28)
Diels–Alder mechanism.
These observations led us to nal of NO-NIF radicals was not symmetrical, as shown in
hypothesize that NO-NIF was responsible for the additional Fig. 3, but asymmetrical, it was evident that the NO-NIF rad-
28,54)
beneficial effect of nifedipine through its antioxidant activity. ical species was located in the membrane environment.
However, it was unclear which species, NO-NIF or NO-NIF Nifedipine selectively accumulates in the membrane, and
radicals, contributed to its antioxidative activity or how NO- it was reported that the membrane-based partition coefficient
55)
NIF exerts its effects. Therefore, in the present study, we in- was 1000.
Although the retention time of NO-NIF
vestigated the antioxidant activities of nifedipine, NO-NIF, (3.4 min) was shorter than that of nifedipine (4.3 min) in the
and mixtures of NO-NIF with various unsaturated fatty octadecylsilyl (ODS) column system, it is conceivable that
19)
acids.
NO-NIF is still lipophilic and locates in the membrane,
As reported previously, NO-NIF gave three-line EPR sig- where it is changed to NO-NIF radicals by reaction with un-
28)
nals when mixed with unsaturated fatty acids, whereas it saturated fatty acids in the lipid bilayer, because cell mem-
56)
did not when mixed with a saturated fatty acid (Fig. 3). branes contain 40 mol% unsaturated fatty acids.
These phenomena indicated that the NO-NIF radical was re- When nifedipine was mixed with HUVECs instead of NO-
sponsible for the interaction of its nitroso group with unsatu- NIF, no EPR signal was observed (data not shown), although
rated double bounds of lipids through pseudo-Diels–Alder NO-NIF radical production from nifedipine in liver tissue
2
3,28,47—50)
23)
reactions.
In addition, the signal intensities of NO- was previously reported. The discrepancy between our data
NIF radicals with unsaturated fatty acids were all the same, and previous works suggest that hepatic enzymes contribute
23)
suggesting that the formation of NO-NIF radicals was unre- to the generation of NO-NIF from nifedipine.
lated to the number of unsaturated double bonds. In addition,
Incidentally, it was reported that nifedipine did not inhibit
18,21)
the NO-NIF radical was stable
enough to measure by ROS production from cells stimulated by 7-ketocholesterol
43)
EPR spectrometry, and gradually increased with time as 6% or TNF-a in endothelial cells. When HUVECs were incu-
of NO-NIF was converted to NO-NIF radicals over a 4-d in- bated with 10 mM LY83583 for 5 min, formation of
cubation when 1 mM NO-NIF was mixed with 1 mM linoleic DMPO/˙OH adducts was observed (Fig. 7D). This EPR sig-
acid (Fig. 4). In the previous study, NO-NIF showed a potent nal was decreased by the addition of SOD (Fig. 7E), suggest-
ꢁ
scavenger effect towards ABAP-derived alkylperoxyl radicals ing that O˙2 was produced from HUVECs in the culture
27)
by electron transfer reaction. However, whether NO-NIF media after LY83583 stimulation. We did not detect the
ꢁ
mixed with a fatty acid acts as a free radical scavenger was DMPO/O˙2 adduct from HUVECs because it decomposed
57)
unknown. Therefore, we studied the radical scavenging activ- rapidly to the DMPO/˙OH adduct in the cell system. In this
ity of reaction mixtures containing NO-NIF with or without study, NO-NIF but not nifedipine decreased the formation of
35)
fatty acids by DPPH assay and LMB assay. When NO-NIF DMPO/˙OH adducts in a concentration-dependent manner
was mixed with DPPH, little scavenging activity was ob- (Figs. 7F—I), suggesting that NO-NIF and/or NO-NIF radi-
ꢁ
served (Fig. 5). Nifedipine did not interact with DPPH radi- cals may contribute to the suppression of O˙2 generation
43)
cals as reported previously (data not shown). However, from HUVECs stimulated by LY83583. We could not distinꢁ-
mixtures containing NO-NIF and unsaturated fatty acids sig- guish which species contributes to the suppression of O˙
2
nificantly scavenged DPPH radicals, suggesting that radi- production by HUVECs stimulated with LY83583 because
cal–radical interactions might be involved. In the other hand, NO-NIF gradually changed to NO-NIF radicals during incu-
the mixtures containing NO-NIF and unsaturated lipids have bation (Fig. 6), although our present data suggest that NO-
51)
ꢁ
electrons to offer electrophiles.
NIF has the ability to reduce O˙ -mediated oxidative stress
2
The scavenging activity against 13-HPODE increased in in endothelial cells independent of calcium channel blockage
the order of nifedipine, NO-NIF, and NO-NIF plus linolenic activity.
acid (Fig. 2). 13-HPODE is a significant component of ox-
Finally, we investigated whether NO-NIF protected cul-
idatively-modified low-density lipoprotein (LDL) and has tured cells from the oxidative stress induced by Fe-NTA
5
2)
been shown to be present in atherosclerosis lesions.
complex. We chose the Fe-NTA complex because we previ-
ꢄ
3
Natarajan et al. reported that 13-HPODE not only activates ously demonstrated that a solution of the Fe -NTA complex
the mitogen activated protein kinases (MAPKs) (extracellular changed to the Fe -NTA complex during storage under light
signal-regulated kinase (ERK), p38, and c-Jun N-terminal ki- and that the Fe -NTA complex produced ROS including
nase (JNK)), but nuclear factor-kappa B (NF-kB) DNA O˙ and hydroxyl radical (˙OH). Likewise, Fe -NTA in-
binding activity and Ras as well as vascular cell adhesion duced oxidation of linoleic acid in the presence of a linoleic
molecule-1 (VCAM-1) promoter activation, and these phe- acid hydroperoxide (LOOH) such as 13-HPODE as fol-
nomena were blocked by antioxidants. This led us to pre- lows.
dict that NO-NIF plus unsaturated fatty acids would be able
to suppress the inflammation-related cell signaling that pro-
motes atherosclerosis, which NO-NIF and nifedipine do not.
Next, we investigated whether the NO-NIF interacts with
living cell systems to produce NO-NIF radicals. In these ex-
periments, we chose HUVECs because of endothelial cell
dysfunction leads to the development of atherosclerosis, and
it is known that nifedipine has pleiotropic effects on these
2
ꢄ
2
ꢄ
ꢁ
58)
2ꢄ
2
53)
59)
3
ꢄ
2ꢄ
Fe -NTA→Fe -NTA (photo irradiation)
2
ꢄ
3ꢄ
ꢁ
Fe -NTAꢄO →Fe -NTAꢄO2˙
2
ꢁ
ꢄ
2
O˙ ꢄ2H →H O ꢄO
2 2 2
2
2
ꢄ
3ꢄ
ꢁ
Fe -NTAꢄH
O
→Fe -NTAꢄOH ꢄ˙OH
2
2
2
ꢄ
3ꢄ
ꢁ
Fe -NTAꢄLOOH→LO˙ꢄFe -NTAꢄOH
37)
cells. Apparent EPR signals of NO-NIF radicals appeared
after 10 min incubation, then gradually increased for 6 h, and
LO˙ꢄLH→LOHꢄL˙ (initiation reaction)

February 2011
213
Therefore, it was conceivable that NO-NIF diminished the
cytotoxicity induced by Fe-NTA because, as we demon-
strated, NO-NIF possessed scavenging activity due not only
to lipid hydroperoxide (Fig. 2) but also ROS (Fig. 7).
J. Pharm. Pharmacol., 39, 1044—1046 (1987).
7) Nunez-Vergara L. J., Bollo S., Fuentealba J., Sturm J. C., Squella J. A.,
Pharm. Res., 19, 522—529 (2002).
8) Stasko A., Brezova V., Biskupic S., Ondrias K., Misik V., Free Radic.
Biol. Med., 17, 545—556 (1994).
1
1
As expected, pretreatment with NO-NIF completely pre- 19) de Vries H., Beijersbergen van Henegouwen G. M., J. Photochem.
vented the Fe-NTA-induced LDH release from PC12 cells,
but nifedipine did not attenuate the cell injury, suggesting
that the antioxidative activity of NO-NIF is superior to that
of its mother compound, nifedipine. In 1992, Mak et al., re-
Photobiol. B, 43, 217—221 (1998).
0) Bauer V., Rekalov V. V., Juranek I., Gergel D., Bohov P., Br. J. Pharma-
col., 115, 871—874 (1995).
2
2
1) Diaz-Araya G., Godoy L., Naranjo L., Squella J. A., Letelier M. E.,
Nunez-Vergara L. J., Gen. Pharmacol., 31, 385—391 (1998).
ported that lipophilic DHPs exhibited antioxidant activity 22) Ohkubo T., Noro H., Sugawara K., J. Pharm. Biomed. Anal., 10, 67—
and that the mechanism might be mediated by membrane
70 (1992).
14,60)
23) Fujii H., Berliner L. J., Magn. Reson. Med., 42, 691—694 (1999).
4) Sanguinetti M. C., Kass R. S., Biophys. J., 45, 873—880 (1984).
5) Gurney A. M., Nerbonne J. M., Lester H. A., J. Gen. Physiol., 86,
lipid antiperoxidation.
However, our results indicate that
2
2
nifedipine seems to have less potent antioxidative activity
14,61)
than other DHPs.
Furthermore, recent studies demon-
3
53—379 (1985).
strated that azelnidipine, a kind of DHP, inhibited or scav- 26) Hayase N., Inagaki S., Abiko Y., J. Pharmacol. Exp. Ther., 275, 813—
4
3,62,63)
8
21 (1995).
enged ROS in both in vivo and in vitro systems.
2
7) Yanez C., Lopez-Alarcon C., Camargo C., Valenzuela V., Squella J. A.,
Nunez-Vergara L. J., Bioorg. Med. Chem., 12, 2459—2468 (2004).
8) Mi sˇ ík V., Stasko A., Gergel D., Ondrias K., Mol. Pharmacol., 40,
435—439 (1991).
In conclusion, we found that NO-NIF accumulated in the
cell membrane and was partially but gradually converted to
stable NO-NIF radicals. These compounds may participate in
2
inhibition of lipid peroxidation of the lipid membrane by di- 29) Akai K., Tsuchiya K., Tokumura A., Kogure K., Ueno S., Shibata A.,
Tamaki T., Fukuzawa K., Free Radic. Res., 38, 951—962 (2004).
0) Mizuta Y., Masumizu T., Kohno M., Mori A., Packer L., Biochem.
Mol. Biol. Int., 43, 1107—1120 (1997).
1) Souchard J. P., Barbacanne M. A., Margeat E., Maret A., Nepveu F.,
Arnal J. F., Free Radic. Res., 29, 441—449 (1998).
rect interaction with lipid hydroperoxide and dismutation of
3
lipid radicals, and may suppress ROS production. Because
23)
NO-NIF is one of the metabolites of nifedipine in vivo and
3
its intrinsic toxicity was less than or equal to that of nifedip-
ine in PC12 cells (Fig. 8), we expect that NO-NIF may be a 32) Auerbach B. J., Kiely J. S., Cornicelli J. A., Anal. Biochem., 201,
3
75—380 (1992).
3) Somers P. K., Medford R. M., Saxena U., Free Radic. Biol. Med., 28,
532—1537 (2000).
4) Uchiyama M., Suzuki Y., Fukuzawa K., Yakugaku Zasshi, 88, 678—
83 (1968).
candidate for an antioxidant drug to protect the cell mem-
brane.
3
3
3
1
Acknowledgment The authors wish to thank Eri Kitayama for her tech-
nical assistance. This work was partially supported by Grant from regional
innovation cluster program.
6
5) Gorinstein S., Martin-Belloso O., Katrich E., Lojek A., Ciz M.,
Gligelmo-Miguel N., Haruenkit R., Park Y. S., Jung S. T., Trakhten-
berg S., J. Nutr. Biochem., 14, 154—159 (2003).
References
36) Hasegawa T., Bando A., Tsuchiya K., Abe S., Okamoto M., Kirima K.,
Ueno S., Yoshizumi M., Houchi H., Tamaki T., Biochim. Biophys.
Acta, 1670, 19—27 (2004).
1)
Simon A., Gariepy J., Moyse D., Levenson J., Circulation, 103,
949—2954 (2001).
2
2)
Poole-Wilson P. A., Lubsen J., Kirwan B. A., van Dalen F. J., Wagener
G., Danchin N., Just H., Fox K. A., Pocock S. J., Clayton T. C., Motro
M., Parker J. D., Bourassa M. G., Dart A. M., Hildebrandt P., Hjalmar-
son A., Kragten J. A., Molhoek G. P., Otterstad J. E., Seabra-Gomes
R., Soler-Soler J., Weber S., Lancet, 364, 849—857 (2004).
37) Yamagishi S., Nakamura K., Matsui T., Curr. Med. Chem., 15, 172—
177 (2008).
38) Kitakaze M., Asanuma H., Takashima S., Minamino T., Ueda Y.,
Sakata Y., Asakura M., Sanada S., Kuzuya T., Hori M., Circulation,
101, 311—317 (2000).
3
)
)
Yui Y., Sumiyoshi T., Kodama K., Hirayama A., Nonogi H., Kanma-
tsuse K., Origasa H., Iimura O., Ishii M., Saruta T., Arakawa K.,
Hosoda S., Kawai C., Hypertens. Res., 27, 449—456 (2004).
Shinoda E., Yui Y., Kodama K., Hirayama A., Nonogi H., Haze K.,
Sumiyoshi T., Hosoda S., Kawai C., Hypertension, 45, 1153—1158
39) Lichtlen P. R., Hugenholtz P. G., Rafflenbeul W., Hecker H., Jost S.,
Deckers J. W., Lancet, 335, 1109—1113 (1990).
40) Matsumori A., Nunokawa Y., Sasayama S., Life Sci., 67, 2655—2661
(2000).
41) Fukuo K., Yang J., Yasuda O., Mogi M., Suhara T., Sato N., Suzuki T.,
Morimoto S., Ogihara T., Circulation, 106, 356—361 (2002).
42) Lupo E., Locher R., Weisser B., Vetter W., Biochem. Biophys. Res.
Commun., 203, 1803—1808 (1994).
43) Naito Y., Shimozawa M., Manabe H., Nakabe N., Katada K., Kokura
S., Yoshida N., Ichikawa H., Kon T., Yoshikawa T., Eur. J. Pharmacol.,
546, 11—18 (2006).
4
(2005).
5
6
)
)
Ding Y., Vaziri N. D., J. Pharmacol. Exp. Ther., 292, 606—609 (2000).
Taddei S., Virdis A., Ghiadoni L., Sudano I., Salvetti A., Curr. Hyper-
tens. Rep., 2, 64—70 (2000).
Berkels R., Egink G., Marsen T. A., Bartels H., Roesen R., Klaus W.,
Hypertension, 37, 240—245 (2001).
Brovkovych V. V., Kalinowski L., Muller-Peddinghaus R., Malinski T.,
Hypertension, 37, 34—39 (2001).
Verhaar M. C., Honing M. L., van Dam T., Zwart M., Koomans H. A.,
Kastelein J. J., Rabelink T. J., Cardiovasc. Res., 42, 752—760 (1999).
7)
8)
9)
44) Sugawara H., Tobise K., Onodera S., Biochem. Pharmacol., 47, 887—
892 (1994).
45) Janero D. R., Burghardt B., Biochem. Pharmacol., 38, 4344—4348
(1989).
1
0) Sugano M., Tsuchida K., Makino N., J. Cardiovasc. Pharmacol., 40,
46—152 (2002).
46) Ondrias K., Misik V., Stasko A., Gergel D., Hromadova M., Biochim.
Biophys. Acta, 1211, 114—119 (1994).
1
1
1) Yamasaki K., Aoki M., Makino H., Hashiya N., Shimizu H., Ohishi
47) Sullivan A. B., J. Org. Chem., 31, 2811—2817 (1966).
48) Floyd R. A., Soong L. M., Stuart M. A., Reigh D. L., Arch. Biochem.
Biophys., 185, 450—457 (1978).
49) Mason R. P., Kalyanaraman B., Tainer B. E., Eling T. E., J. Biol.
Chem., 255, 5019—5022 (1980).
50) Sammartano L. J., Malejka-Giganti D., Chem. Biol. Interact., 77, 63—
79 (1991).
51) Abe S., Kirima K., Tsuchiya K., Okamoto M., Hasegawa T., Houchi
H., Yoshizumi M., Tamaki T., Chem. Pharm. Bull., 52, 186—191
(2004).
M., Ogihara T., Morishita R., J. Hum. Hypertens., 18, 701—705
(
2004).
2) Himmel H. M., Whorton A. R., Strauss H. C., Hypertension, 21,
12—127 (1993).
3) Sugawara H., Tobise K., Kikuchi K., Hypertens. Res., 19, 223—228
1996).
4) Mak I. T., Zhang J., Weglicki W. B., Pharmacol. Res., 45, 27—33
2002).
1
1
1
1
(
(
15) Yamagishi S., Nakamura K., Med. Hypotheses., 68, 565—567 (2007).
16) Majeed I. A., Murray W. J., Newton D. W., Othman S., Al-Turk W. A.,
52) Folcik V. A., Cathcart M. K., J. Lipid Res., 35, 1570—1582 (1994).

2
5
5
5
5
5
5
14
Vol. 59, No. 2
Fukuzawa K., Biochim. Biophys. Acta, 1725, 111—119 (2005).
59) Fukuzawa K., Fujii T., Mukai K., Arch. Biochem. Biophys., 290, 489—
496 (1991).
60) Mak I. T., Boehme P., Weglicki W. B., Circ. Res., 70, 1099—1103
(1992).
61) Mak I. T., Boehme P., Weglicki W. B., Biochem. Pharmacol., 50,
1531—1534 (1995).
62) Yamagishi S., Inagaki Y., Nakamura K., Imaizumi T., J. Cardiovasc.
Pharmacol., 43, 724—730 (2004).
63) Jinno T., Iwai M., Li Z., Li J. M., Liu H. W., Cui T. X., Rakugi H.,
Ogihara T., Horiuchi M., Hypertension, 43, 263—269 (2004).
3) Natarajan R., Reddy M. A., Malik K. U., Fatima S., Khan B. V., Arte-
rioscler. Thromb. Vasc. Biol., 21, 1408—1413 (2001).
4) Borbat P. P., Costa-Filho A. J., Earle K. A., Moscicki J. K., Freed J. H.,
Science, 291, 266—269 (2001).
5) Mason R. P., Trumbore M. W., Biochem. Pharmacol., 51, 653—660
(
1996).
6) Van Blitterswijk W. J., De Veer G., Krol J. H., Emmelot P., Biochim.
Biophys. Acta, 688, 495—504 (1982).
7) Halliwell B., Gutteridge J. M. C., Clarendon Press, 1989, 47—53
(1989).
8) Tsuchiya K., Akai K., Tokumura A., Abe S., Tamaki T., Takiguchi Y.,